Neutron porosity logging and core porosity measurements in the Beauvoir granite, Massif Central Range, France

Abstract

A large suite of geophysical logs have been run in the Beauvoir granite. The drillhole (900 m deep), first target of the French Deep Geology programme, is located in the Hercynian bedrock of Echassières in central France (Massif Central Range). After geochemical and petrological studies, the batholith was used for experiments pertaining to the storage of radioactive wastes. With its low porosity, its weak fracturing and its high homogeneity, the Beauvoir granite was chosen for the analysis of the relationship between logged data and the properties measured in the core. The study focused on neutron porosity and core water porosity. The Beauvoir granite has a total free water porosity of around 2% (average value of 54 core samples of rock mass) and an average neutron porosity of around 10%. We show that the origin of this significant difference is related to the neutron matrix effect of the granite. This phenomenon is partly due to the slowing-down effect of the combined water of clays and micas but also to the neutronic capture effect linked with the relatively high lepidolite (lithium mica) content of the granite. The Li 2O content controls 85% of the granite macroscopic capture cross-section. These two factors represent around 75% of the global neutron porosity of the Beauvoir granite. They have to be taken into consideration to get representative water contents of a low-porosity igneous rock from a neutron porosity log. Further investigations also demonstrated the necessity of choosing a better adapted neutron tool calibration for crystalline rocks. Instead of a standard calibration with limestone blocks, a calibration in granite blocks was simulated in order to obtain a better evaluation of the global neutron response of the granite. Then, by correcting this new neutron porosity for the matrix effect, it has been possible to determine water contents in accordance with laboratory water porosity values measured on core samples (2% average porosity). This experimental study shows that geophysical logging must be adapted to the type of rock and that the geological context (mineralogy, geochemistry) has to be taken into account before running the tool in the borehole. The technique could help determine if neutron porosity data are revelant in other granitic or crystalline rocks.